What's Under the Skin? Estimating Swine Body Condition
Researchers have developed PigFormer, a novel system designed to accurately estimate swine body condition using RGB-D cameras. This two-stage system processes depth frames to predict backfat thickness and loin muscle depth, offering a more scalable and automated alternative to traditional manual methods. PigFormer achieved a mean absolute error of 2.43 mm for backfat and 3.87 mm overall on a dataset of 319 swine instances, outperforming existing baseline models. AI
IMPACT Provides a more accurate and automated method for monitoring livestock health, potentially improving agricultural efficiency.